Help me figure out growth path?


#1

Hello to all who were great enough to come read my post.

I am putting a business plan together and have chosen dreamhost to initially host my website. I am trying to forecast as best as I can the growth plan. I will be starting with the basic shared server plan (the $8.95/month) and need to know how best to gauge the length at which this plan will need to be upgraded to a VPS? Such as something like, “20,000 visits per day” or “200 page requests per second”, etc. Are there a good set of rules or guidelines that will help me do this? Also, I think it will make sense to step through the various VPS options as well, mostly the RAM I pay for and what triggers will cause me to step that up until I eventually colo some rack space and put up servers of my own online (everyone’s dream right? to get this big).

As far as information on the website goes, it is not solid. I am either going to go with something like drupal or concrete5 (thanks to DH Newsletter for that!) or creating the entire site from scratch. It will be a blog (wordpress) + main site (something like a cross between linkedin and facebook where users have profiles and interact with one another on a more business level) + A sort of specific Forum/BB. Also, SSL in all the right places as security will not be taken lightly.

So with that said, any advice or tips anyone has to help make this as realistic as an estimate can get, I am highly grateful. Thanks for stopping by!


#2

Sadly, there’s no real “Rule of Thumb” regarding resource usage forecasting. The fact is with dynamic sites that there are a multitude of differing variables for each and every website design (indeed, each and every page!). If you ever see statistics posted regarding hits versus resource usage they are at a minimum simply a “best guess” and really can not be used - even as a guideline.

In order to lower resource usage use static pages whenever possible. If your pages are all dynamic in nature, choose a good caching mechanism that will help limit the amount of database hits imposed during visits.